P. Kyle House, PhD., Surficial Geologic Mapping Specialist / Geomorphologist
Figure 15. Aerial photograph of river kilometers 37.5–39.5 showing presumed initial location of West Crater lava dam. The <st1:place st="on"><st1:placename st="on">Owyhee</st1:placename> <st1:placetype st="on">River</st1:placetype></st1:place> flows to the north in this image. The red dashed ellipse marks the location of a possible spillway in the West Crater dam and/or the initial point of breaching. The asterisk denotes the location of the sediment lens depicted in Figure 16. Ryegrass Creek defines the western edge of the Saddle Butte lava flow. The West Crater lava flow entered the <st1:place st="on"><st1:placename st="on">Owyhee</st1:placename> <st1:placetype st="on">Canyon</st1:placetype></st1:place> by flowing down the Bogus Creek canyon. Enough lava was supplied to the <st1:place st="on"><st1:placename st="on">Owyhee</st1:placename> <st1:placetype st="on">Canyon</st1:placetype></st1:place> that the dam was able to grow upstream towards Ryegrass Hot Spring.
Figure 16. Photograph of outcrop relations at river kilometer 39.25. The Saddle Butte lava appears to underlie the sediment lens in the center of the photograph. The head of the West Crater lava dam overlies the sediment lens and the Saddle Butte lava. Foreset beds of pillow lavas and hyaloclastite dip down to the right in this image. Several different passage zones (indicated by PZ) indicate the lake level rose during construction of the dam. The location of the paleovalley wall discussed in Figure 15 is indicated by a dashed ellipse. The lens of sediment interpreted to be paleo-Ryegrass Creek is visible in the center of the photograph.
Oooohh Boy, Here are some ramblings for you to chew on, some of which I have been chewing on for some time (years). Dr. Jerque’s previous comments are in italics. Sorry to post a dry reply without any figures--I am having trouble pulling them into the blog--will try on Sunday. Figures 2, 15, and 16 are pretty relevant to the discussion.
The upper West Crater lava in the presumed abutment on river right does not have lava-delta deposits (wtf?). In the field, it looks like a dry flow. Why would water have not backed up in this area during the blockage? The contact of young WC on Old WC marks the perimeter of the 3400 ft lake. I suppose this part could have been dry initially as the flow continued in a generally downstream direction....
Yes, the youngest WC (above the 3400 ft contour) does not appear to have any evidence of lava-water interaction. But, the uppermost flow units immediately below 3380 ft surface do transition quickly (within a few meters) into a lava pillow delta via a passage zone (see Figure 2 in my thesis). I can think of at least two options for why the lava above the 3400 ft contour doesn’t have any pillows: 1) Lava was entering the river canyon at a much faster rate than the river discharge (and the rate of lake rise) such that the lava dam growth outpaced the rise of the reservoir; 2) the level of the reservoir stabilized at ~3370 ft (perhaps because a stable spillway developed adjacent to Pruitt’s Castle or because the dam was porous). In either case, if the full discharge of the river could seep through or sneak around the dam, the crest of the dam could grow uninhibited and not be within the reach of the water.
<o:p> </o:p>
Option 1 is complicated by the fact that we know that the rate of dam construction and lake level rise was semi-episodic because of the presence of multiple passage zones (and accompanying subaerial lava) preserved within the dam.
<o:p> </o:p>Option 2 could work because portions of the dam could be porous as observed by Crow et al. (2008), a spillway could have been eroded into the ridge of Tertiary that makes Pruitt’s Castle (near the white star in figure 15 of thesis) contemporaneously with dam construction and lake filling, or some combination of both. In regards to the porosity of the dam, our friendly neighborhood p-mag expert and I discussed this during the recent trip. I had originally conceptualized that the pillow lava deltas would be rather porous but our p-mag expert pointed out that they are likely rather well-consolidated for several reasons. For example, as the deltas form, they are sort of self-packing—a variety of clast sizes are settling and snuggling together as they tumble down and more material is added from above. In addition, some large clasts (pillows) could still be somewhat plastic and deform to fit the space provided them, almost welding together. Depending on the dissolved gases in the lava, its temperature, and the ambient pressure at the locus of emplacement (and maybe some other parameters), the crust of the growing lobes and pillows of lava fractures into tiny glassy quenched bits (hyaloclastite) that serve to fill any interstices in the delta. This hyaloclastite can dominate the delta by volume, leaving the delta matrix supported, and when the hyaloclastite devitrifies, lots of clay minerals are produced further reducing the hydraulic conductivity of the dam (an turning some portions of the dam tan-orange in color—see photo). In contrast, Crow et al. (2008) identified actual cinders in the presumed abutments of some of their lava dams and they even called some dams “sieves” rather than dams. Depending on how these pyroclastics in the <st1:place st="on">Grand Canyon</st1:place> were emplaced, they could provide the necessary porosity to accommodate the discharge and stabilize the lake height. How about that? Are there other models you can think of to explain the outcrops?
<o:p> </o:p>
In regards to Dr. Jerque’s questions about the timing of individual incursions of lava into the river and the total lifespan of the obstruction created, I think the dams are built quickly. The vents supplying lava to the intracanyon lava flows are monogenetic and probably have a life span of months, years, or perhaps tens of years, but not hundreds of years. From my understanding of Snake River Plain volcanism and the experience of those such as our p-mag expert, these lava flows could easily erupt, flow across the uplands, and build a dam in a few months or maybe several years. During that time, individual pulses of lava (flow units, cooling units, surges, etc.) added to the obstructions created by the first lavas in a tug of war battle with the rising lakes. The multiple (and rising in elevation) passage zones at Weeping Wall and WC at tell us this. The resulting dam is so geologically instantaneous that it makes sense to me to model it mostly as a single event. The details of the passage zone elevations, relative amounts of subaerial vs subaqueous lava, and volume of hyaloclastite tell us some of the juicy details of the event but in terms of the ~2 Ma history of the river available to us to model I would consider it one event. Even in the case of the SB dam, where there are two clearly different advances into the canyon (that potentially could be separated by a lot of time) we do nto see any different in age with the p-mag. We also do not see any fluvial deposits intercalated within the dam architecture that would suggest a long time interval (1000s to 10000s of years). What do you think?
<o:p> </o:p>
One of my thesis’ objectives was to try to add data to, and refine, lava dam emplacement and breaching models (and the associated hazards) by trying to study how the rate of lava effusion into a river interacts with the river’s discharge and channel morphology to influence the structure and stability of lava dams. This objective was often overshadowed by the larger objective of just trying to figure out what he-ack is going on out there and distinguish the lava flows from one another but I do think that there is enough data to address the matter in the paper I am putting together on the lava flows.
Cheers,
Spud
Here is a thought-provoking image that I created in Global Mapper this evening (click on it to enlarge). It shows the extent of a dam with a crest elevation of 3400 feet, which is the approximate elevation of the Saddle Butte 2 and West Crater Lava Dams (right, I couldn't get the blue from extending downstream, but you get the point). I have sketched in some likely dam locations...the Saddle Butte Dam is obvious because both abutments still exist. The West Crater dam, however, is a little harder to guess. The right (east) abutment is pretty obvious, but the left (west) one is not. In both cases, as we know, Rome Valley was inundated. Duh (in hindsight)
Other points:The upper West Crater lava in the presumed abutment on river right does not have lava-delta deposits (wtf?). In the field, it looks like a dry flow. Why would water have not backed up in this area during the blockage? The contact of young WC on Old WC marks the perimeter of the 3400 ft lake. I suppose this part could have been dry initially as the flow continued in a generally downstream direction....In the case of each dam, the damming event that is most well preserved 'immediately' post-dates an intracanyon flow from the same vent. We don't have any direct evidence that these flows dammed the river...but they certainly may have, or did they? wtf?Is it of any interest to the modeling and general conceptualization of the processes that the lava events that clearly dammed the river were damming a river that may still have been in the process of dealing with the previous lava incursion from the same vent? (thus not so long ago, relative to the damming event).Any thoughts? Break down and post some comments or, maybe, your own diagram....
Ok. So I recently learned that ESRI actually provides a kml-making tool in the toolbox. That is handy. I haven't yet figured out how to create a clean version to distribute, but this screen shot should provide some perspective on where this can go.
I used Globalmapper (more on that later) to quickly create this shaded relief image of the study area. Are most of the lava platforms that are so evident in this image related to lava-dams? The various volcanic structural forms in this image are really very striking. If it turns out that all or most of these features reflect late Neogene lava-water interaction, this is a fairly significant landscape-forming process in this region...the distribution of these features on the landscape record the progressive incision of the Owyhee River over a very long time, no?
Thus far, I have spent 10s of hours on the map. As a consequence, the correlation diagram has evolved significantly. The colors in the snippet do not match the map yet, and some of the unit labels have changed. Also, it is likely that some units will be added and some will be dropped as the project continues. All part of my mapping process, and is often maddening to my colleagues. Nonetheless, check it out and go with the flow.
Made some decent strides on the map in THIG today. Check it out. Pretty much the most complex area I have mapped in a while. Thankfully, I don't have to parse out the bedrock.
| Stations (site specific data) | |||
| [kind] O = generic observation | |||
| [kind] A = age | |||
| [kind] G = graphic data | |||
| [kind] R = sample sites | |||
| [kind] Y = analytical | |||
| Age categories (prefix 1) | |||
| a = Argon-Argon | |||
| r = radiocarbon | |||
| t = tephrochronologic | |||
| c = cosmogenic | |||
| Graphic data categories | |||
| p = photograph | |||
| s = sketch | |||
| Sample site categories | |||
| r = rock | |||
| s = sediment | |||
| t = tephra | |||
| Analytical categories | |||
| f = fluvial transport direction | |||
| g = fluvial gravel lag | |||
I am heading out to a different field area, so I thought I would leave you with an interesting shot while I take a week off from the Owyhee. This is a picture of the Bogus lava outcrop at Rinehart Canyon. Here, the Bogus Rim basalt overlies a considerably thicker pile of 'older Bogus lavas'. The tapered marginal pinch of the Bogus Rim unit is hilarious in its clarity. This contrasts with the weirdness associated with the seemingly detached block surrounded by sediments along the base of the lower lavas, which have an otherwise obvious lower contact. Of equal interest is the scale-providing 'surplus' farm equipment on the slope. Thanks to Cooper for the inset detail.
NBMG recently hired a new chief cartographer with an extensive background in ArcGIS. As a consequence, our new approach to geodatabases is very comprehensive. I am climbing a relatively steep learning curve as I figure it out. One interesting aspect is that the new structure can pull a correlation diagram designed in excel (of all things!) right into a map layout. Any changes to labels and colors on the map are automatically updated in the correlation diagram.
Thus, instead of just mapping nonstop since we left Bend, I have been tackling the basic logistical issues associated with organizing the map data (see previous posts on unit labels and line types). So, today I tackled the correlation diagram. Please look this over and provide any comments you may have. Typically this type of chart evolves as the mapping progresses, but the first stab has to be decent. If you want to edit the spreadsheet directly, let me know and I will forward it. Note that I have added a chronological component to the landslide units. Not sure how easy it will be to divide, but decided the option was important to have.
